Prophylactic efficacy of lithium versus carbamazepine in treatment-naive bipolar patients

Background: Alternatives to lithium for prophylactic treatment of patients with bipolar affective disorders are increasingly being advocated. However, trials comparing lithium with alternatives are scarce and often biased. Method: We studied 94 patients with at least 2 episodes of bipolar disorder (DSM-III-R) during the previous 3 years who were in remission at entry into the study. Treatment with lithium or carbamazepine had not exceeded a total of 6 months during their lifetime. Patients were randomly assigned to carbamazepine or lithium at entry into the 2-year double-blind study or during the acute index episode previous to entry into the study. No concurrent antipsychotics or antidepressants were allowed. Results: On lithium treatment, 12/44 patients developed an episode, compared with 21/50 on carbamazepine treatment. Episodes on lithium treatment occurred almost exclusively during the first 3 months of the trial. Carbamazepine carried a constant risk of an episode of about 40% per year. Efficacy of lithium was superior to that of carbamazepine in patients with a (hypo)manic index episode that had not been treated with study drug during the index episode (p < .01) and also in patients with prior hypomanic but no manic episodes (p < .05). The proportion of patients who dropped out was slightly higher among those taking lithium (16/44) compared with those taking carbamazepine (13/50), resulting in 16/44 patients (36%) on lithium treatment completing the 2 years with no episode, compared with 16/50 (32%) on carbamazepine treatment. Conclusion: Lithium appears to be superior in prophylactic efficacy to carbamazepine in bipolar patients not previously treated with mood stabilizers. Our results should reinforce efforts to put and maintain such patients on treatment with lithium.